Bundles
Curated themes. Each starts from anchor papers and grows every night from the works that cite them.
Active
A| Bundle | Papers | Green | Yellow | Orange | Red |
|---|---|---|---|---|---|
| AI for science and agentic research Agents that plan, run and write up research, and where they still fall short of experts. | 34 | green: 21 | yellow: 6 | orange: 0 | red: 7 |
| Production agentic workflows and evaluation What makes an agentic workflow fit for daily operation: evaluation beyond task success, deployment practice, and the failure modes of autonomous agents. | 8 | green: 4 | yellow: 0 | orange: 3 | red: 1 |
| RAG reliability and evaluation Retrieval-augmented generation and how to evaluate it: retrieval quality, citation quality, factuality. Carries the verification discipline in act 2. | 29 | green: 15 | yellow: 6 | orange: 3 | red: 5 |
| Research integrity and LLMs What LLMs do to the evidence base: illusions of understanding, fabricated references, and measurable LLM text in papers and peer review. Carries act 2. | 34 | green: 21 | yellow: 5 | orange: 1 | red: 7 |
| Supervision and doctoral education Generative AI in doctoral research and supervision. Thin: needs more papers before it carries act 3. Includes the jagged frontier and human task formulation. | 33 | green: 14 | yellow: 11 | orange: 4 | red: 4 |
| Uppsala: knowledge infrastructure Uppsala has built knowledge infrastructure before. Linnaeus to genomics. | 4 | green: 1 | yellow: 0 | orange: 2 | red: 1 |
Columns:GreenYellowOrangeRed
Topic packs, switched off
BLoaded on top of the core for a single occasion; they keep growing while off.
- DDLS pack: evolution and biodiversity · 34 papers